Hi Bin,
On 25 February 2015 at 00:10, Bin Meng <bmeng...@gmail.com> wrote:
> Hi Simon,
>
> On Wed, Feb 11, 2015 at 9:59 AM, Simon Glass <s...@chromium.org> wrote:
>> Some systems have more than 4GB of RAM. U-Boot can only place things below
>> 4GB so any memory above that should not be used. Ignore any such memory so
>> that the memory size will not exceed the maximum.
>>
>> This prevents gd->ram_size exceeding 4GB which causes problems for PCI
>> devices which use DMA.
>>
>> Signed-off-by: Simon Glass <s...@chromium.org>
>> ---
>>
>> arch/x86/cpu/coreboot/sdram.c | 6 ++++--
>> 1 file changed, 4 insertions(+), 2 deletions(-)
>>
>> diff --git a/arch/x86/cpu/coreboot/sdram.c b/arch/x86/cpu/coreboot/sdram.c
>> index e98a230..9c3ab81 100644
>> --- a/arch/x86/cpu/coreboot/sdram.c
>> +++ b/arch/x86/cpu/coreboot/sdram.c
>> @@ -90,7 +90,8 @@ int dram_init(void)
>> struct memrange *memrange = &lib_sysinfo.memrange[i];
>> unsigned long long end = memrange->base + memrange->size;
>>
>> - if (memrange->type == CB_MEM_RAM && end > ram_size)
>> + if (memrange->type == CB_MEM_RAM && end > ram_size &&
>> + memrange->base < (1ULL << 32))
>
> Can we safely assume no single entry in memrange[] lies across the 4GB
> boundary?
>From what I have seen, yes. There always seems to be a hole in memory
from about 3.3GB to 4GB.
>
>> ram_size = end;
>> }
>> gd->ram_size = ram_size;
>> @@ -108,7 +109,8 @@ void dram_init_banksize(void)
>> for (i = 0, j = 0; i < lib_sysinfo.n_memranges; i++) {
>> struct memrange *memrange = &lib_sysinfo.memrange[i];
>>
>> - if (memrange->type == CB_MEM_RAM) {
>> + if (memrange->type == CB_MEM_RAM &&
>> + memrange->base < (1ULL << 32)) {
>> gd->bd->bi_dram[j].start = memrange->base;
>> gd->bd->bi_dram[j].size = memrange->size;
>> j++;
>> --
>
> Regards,
> Bin
Regards,
Simon
_______________________________________________
U-Boot mailing list
U-Boot@lists.denx.de
http://lists.denx.de/mailman/listinfo/u-boot
